The Journal of Neuroscience, January 1, 1996, 76(1):31-35 mRNA Expression of KIFIA, KIFIB, KIF2, KIF3A, KIF3B, KIF4, KIF5, and Cytoplasmic during Axonal Regeneration

Reiko Takemura,‘,’ Takao Nakata,’ Yasushi Okada,’ Hiroto Yamazaki,’ Zhizeng Zhang,’ and Nobutaka Hirokawa’ ‘Department of Anatomy and Cell Biology, School of Medicine, University of Tokyo, Hongo, Bunkyo-ku, Tokyo 113, Japan, and 20kinaka Memorial Institute for Medical Research, Toranomon, Minato-ku, Tokyo 105, Japan

Mouse brain expresses multiple superfamily brain but not in the adult, were under the detection limit in (KIFs), which are involved in vesicle transport. The expres- both control and regenerating dorsal root ganglion cells. sion of KlFs is developmentally regulated, and both the Because mRNA of neither KIF2 nor KIF4 increased signifi- mRNA and proteins of KIF2 and KIF4 are expressed abun- cantly, the results suggest that the expression of KlFs dantly in the juvenile brain. To elucidate the role of individual during regeneration does not recapitulate the embryonic kinesin superfamily motor proteins during regenerative out- development and support the hypothesis that different series growth of axons, we examined the mRNA expression of of events take place during the regenerative and embryonic KlFlA, KIFlB, KIF2, KIF3A, KIF3B, KIF4, and KIF5 in adult outgrowths of axons. In contrast, mRNA for cytoplasmic mouse dorsal root ganglion cells after sciatic nerve crush. dynein was slightly increased, up to 140%. This is consistent Seven to fourteen days after the nerve crush, the mRNA with the hypothesis that retrograde transport plays critical expression pattern of and p- isotypes roles in regeneration such as the transport of neurotrophic suggested that the regenerative outgrowth of axons was factors. active. At these stages, levels of mRNA for KIFl A, KIFlB, KIF2, KIF3A, KIF3B, KIF4, and KIF5 were .50-80% of control. Key words: K/F; kinesin; cytoplasmic dynein; axon transport; The levels of mRNA for KIF4, which are detected in juvenile regeneration; neuron

In the axon, various membraneorganelles are transported along mains are quite distinct, and they are all derived from distinct the microtubulesboth anterogradely and retrogradely (Hirokawa, . KIFlA is homologousto C. elegans uncl04 and is an 1993). Kinesin and cytoplasmic dynein have been identified as anterograde motor for transport of synaptic vesicle precursors motor proteins for the transport (Brady, 1985; Vale et al., 1985; (Okada et al., 1995a).On the other hand, KIFlB wasshown to be Paschalet al., 1987;Hirokawa, 1993). Kinesin has been shownto an anterogrademotor for mitochondria transport (Nangakuet al., be involved in the anterogrademovement and cytoplasmicdynein 1994).KIF5 appearsto be an isotypeof kinesin heavy chain but is in the retrograde movement. However, within an axon many expressedalmost exclusively in the brain. KIF3A and KIF3B form different populations of membrane are transported, a heterotrimer with associatedproteins (KAP3) and work as an and it is likely that moleculesother than kinesin and cytoplasmic anterograde motor for membranousorganelles (Kondo et al., dynein are involved in the regulation of the complex traffic of 1994; Yamazaki et al., 1995). KIF2 is unique in having its motor membraneorganelles within the axon. domain in the middle of the moleculeand is expressedabundantly Recently, we have cloned membersof kinesin superfamilypro- in juvenile neurons and transports membranousorganelles dis- teins (KIFs) from mouse brain. Five membersoriginally were tinct from synapticvesicle precursors not carried by kinesin heavy cloned (Aizawa et al., 1992;Hirokawa, 1993).Subsequent studies chain and KIF3A/B (Noda et al., 1995). All other membershave revealed that there are at least sevenmembers in the superfamily, motor domainsat the N-terminal end. Multiple membersap- namely, KIFlA, KIFlB, KIF2, KIF3A, KIF3B, KIF4, and KIF5 pearedto be expressedin the sameneuronal cells as shownby in (Kondo et al., 1994; Nangaku et al., 1994; Sekine et al., 1994; situ hybridization and immunocytochemistry,suggesting that dif- Noda et al., 1995; Okada et al., 1995a;Yamazaki et al., 1995). ferent membersplay different roles in axoplasmictransport within They all share the kinesin motor domain, which contains a puta- a singlecell. tive ATP-binding site and a -bindingsite. Other do- The changesin the gene expressionof neurofilament and tu- bulin during axonal regeneration of the mammalianperipheral Received Aug. 28, 1995; accepted Sept. 19, 1995. nerve systemhave been well studied(Hoffman et al., 1987; Hoff- This work was suooorted bv a Grant-in-Aid for a soecial oroiect research from the man and Cleveland,1988; Miller et al., 1989;Wong and Oblinger, Ministry of Educati’dn, Scienle and Culture of Japan’(N.H,j. tie thank Dr. Nicholas J. Cowan for providing us with cDNAs for MP2, M/34, M/35, and NF68; Dr. Yasushi 1990; McKerracher et al., 1993). Particularly, it has been shown Harihara for valuable advice about surgical procedures on rodents, which was critical that the geneexpression of neurofilamentand tubulin recapitulate for starting up the experiments; and Dr. Hitoshi Kanno for valuable advice on Northern blotting procedures. We also thank Ms. Yukako Okamura, Atsuko the developmental pattern during regeneration. Therefore, the Kawaguchi, and Mihoko Watanabe for their excellent technical assistance. systemis suitablefor examiningthe role of certain proteins in the Correspondence should be addressed to Dr. Nobutaka Hirokawa, Department of elongation of axons. To elucidate the role of different motor Anatomy and Cell Biology, School of Medicine, University of Tokyo, 7-3-l Hongo, Bunkyo-ku, Tokyo 113, Japan. proteinsduring regeneration,we examinedthe geneexpression of Copyright 0 1995 Society for Neuroscience 0270-6474/95/160031-05$05.00/O seven membersof kinesin superfamily proteins, KIFlA, KIFlB, 32 J. Neurosci., January 1, 1996, 76(1):31-35 Takemura et al. . KIF Expression and Regeneration

P TUBULIN NF68 II IV I c r c r c r c r

2.4 -

Figure 1. Northern hybridization autoradiograms of mouse DRG RNA probed with p-tubulin isotypes and NF68. Seven days after nerve crush, RNA was isolated from L4 and L5 DRG. The same amount of RNA was loaded for each set of control (c) and regenerating (r) lanes. Each filter was hybridized with probes for class II, IV, or I P-tubulin, or NF68. Ten micrograms of RNA were used for class II /3-tubulin, and 2 pg each for class I and IV /.?- and NF68. The numbers on the Zef indicate the size of the marker RNA in kilobases.

KIF2, KIF3A, K.IF3B, KIF4, and KIF5, and cytoplasmicdynein, in fi-tubulin to standardize the variability of the amount of RNA loaded for mouselumbar dorsal root ganglion cells (DRG) during regener- the control and regenerating lanes. Probes. The 3’-noncoding regions of MP2, MP4, and MP5 and the ation by Northern hybridization. coding region of NF68 were gifts from Dr. N. J. Cowan of the Depart- ment of Biochemistry, New York University Medical Center (Lewis and MATERIALS AND METHODS Cowan, 1985; Lewis et al., 1985). For detection of KIFs, in addition to the Surgical operation. Surgical procedures performed for rats (Hoffman and probes described previously (Aizawa et al., 1992), a 630 bp fragment from Cleveland, 1988) were modified and adopted for mice. Namely, 7-week- the C-terminal tail region of KIFlA, the whole length of KIFlB, a 1.2 and old male C57BL mice weighing -20 gm were treated with sodium a 1.1 kb fragment from C-terminal tail regions of KIF3A and KIF3B were pentobarbital (6 mg/kg) and anesthetized with ether. Then the sciatic used. For the detection of cytoplasmic dynein, a 916 bp fragment of rat nerves were crushed bilaterally at the junction of I+4 and L5 spinal nerves cytoplasmic dynein from 6664 to 7580 (Zhang et al., 1993) was used. by tightly pinching the nerve twice for 30 set using No. 5 forceps. Femoral Quantification. After Northern hybridization, the filters were exposed and obturator nerves also were crushed in experiments in which the mice to Hyperfilm-MP (Amersham, Arlington Heights, IL) preflashed accord- were killed within 7 d after the operation. ing to a published method (Laskey, 1980) with an intensifying screen at RNA isolation. Seven and fourteen days after crushing the nerves, L4 -80°C. The results were quantified with a densitometer (Elscript 400, and L5 DRG were removed from mice and immediately frozen in liquid Hirschmann, Unterhaching, Germany) or imaging plates (BAS 2000, Fuji nitrogen. For each batch of RNA, -80 DRG were pooled and RNA was Film, Tokyo, Japan). isolated from DRG by acid guanidinium thiocyanate-phenol-chloroform extraction (Chomczynski and Sacchi, 1987). Control RNA was isolated in the same manner from L4 and L5 DRG of 7-week-old male C57BL mice. RESULTS Northern hybridization. The same amounts of total RNA obtained from Changes in mRNA expression of neurofilament and the control and regenerating mice were electrophoresed on 1% agarose gels containing formaldehyde and transferred to nitrocellulose filters. tubulin during regeneration in mouse DRG Two to ten micrograms of RNA was used, depending on the sensitivity of Changesin mRNA expressionof NF68 and classI, II, and IV the detection of the probes used. The filters were then hybridized with the /3-tubulins have been well documented in rat lumbar sensory probes labeled with [u-32P]dCTP using the random primer labeling neurons during axonal regeneration (Hoffman and Cleveland, method in a hybridization buffer at 42°C overnight. They were then washed to a final stringency of 0.1 X SSC, 0.1% SDS at 60°C for KIFs, 1 X 1988). Becausewe usedsimilar surgical procedures,we first ex- SSC, 1% SDS at 60°C for p-tubulins, or 1X SSC, 1% SDS at 50°C for 60 aminedwhether similar changeswere seenin mice. As shownin min for cytoplasmic dynein probes. For a set of experiments using RNA Figure 1 and Table 1, the level of expressionof classII tubulin collected 7 d after nerve crush, each filter was rehybridized with class IV mRNA increased(1.4- to 4.7-fold), and that of NF68 mRNA decreased(40-60% of control) between7 and 14 d after crushing Table 1. mRNA expression of /34ubulin and NF68 in regenerating the nerves. These changesare those seen during the phase of mouse DRGs active elongation of axonsin rat. Therefore, the resultssuggested that regenerationwas proceedingsuccessfully in mice during this 14 d after period. 7 d after nerve crush nerve crush mRNA (% of control) (% of control) Changes in mRNA expression of KlFs during regeneration in mouse DRG ARG IP ARG P-tubulin I 132 133 124 We next examined changesin the mRNA expression of the II 470 223 141 kinesin superfamily proteins KIFlA, KIFlB, KIF2, KIF3A IV 106 108 101 KIF3B, KIF4, and KIF5 in regenerating and control mice. Rep- NF68 43 58 62 resentative autoradiogramsare shown in Figure 2, and quantita- tive resultsare shown in Table 2. The Northern blots such as those shown in Figure 1 were quantified by densitometry We first examinedthe changesobserved 14 d after nerve crush. of the autoradiograms (ARG) or by imaging plates (IP) for class I, II, and IV P-tubulins and NF68. The results were expressed as the percent of mRNA expression The relative amount of RNA needed to detect each KIF probe in regenerating DRG compared with control. suggestedthat relative expressionpattern of IUFs in control DRG Takemura et al. l KIF Expression and Regeneration J. Neuroscl., January 1, 1996, 16(1):31-35 33

KIF 2 3A 3B 5

C r C r c r c r c r

4.4- ‘111

PTUB ” l

Figure 2. Northern hybridization autoradiograms of mouse DRG RNA probed with KIFs. Seven days after nerve crush, RNA was isolated from L4 and L.5 DRG. Pairs of RNA were loaded for control (c) and regenerating (r) lanes. Filters were hybridized with probes for KIFlA, KIFZ, KIF3A, KIF3B, and KIFS.Approximately 2 pg wasused for KIFlA andKIF5,5 pg for KIF3A and KIF3B, and 10 pg for KIF2. As a control for the amount of RNA loaded for (c) and (r), all of the filters were rehybridized with probes for class IV P-tubulin, which did not change the expression during regeneration. The numb&son the left indicatethe sizeof the markerRNA in kilobases. are similarto the adult brain. After Northern hybridization using (Aizawa et al., 1992). If the mRNA expressionpattern of KIFs the sameamount of RNA for control and regeneratingDRG, the reverted to the juvenile pattern during regenerationas p-tubulin expressionof mRNA was quantified by densitometry of autora- and NF68, it waslikely that KIF4 mRNA was detected consider- diograms.The resultswere expressedas percent of control. As ing the condition usedfor the hybridization. Therefore, mRNA of shownin Table 2, the levels of mRNA of KIFlA, KIFlB, KIF2, KIF2 and KIF4, both of which are expressedabundantly in juve- KIF3A, KIF3B, and KIFS in regeneratingmice were .50-80% of nile brain (Aizawa et al., 1992; Sekine et al., 1994; Noda et al., control. KIF4 was below the detection limit in both control and 1995), did not increasesignificantly during regeneration.Thus, it regeneratingmice. KIF4 mRNA is detected abundantly in juve- was suggestedthat unlike P-tubulin or NF68, the geneexpression nile brain, but its level is under the detection limit in the adult of KIFs did not recapitulate the developmental pattern during regeneration. TabIe 2. mRNA expression of KIFs and cytoplasmic dynein in We also examined the mRNA isolated 7 d after nerve crush regenerating mouse DRG (Fig. 2, Table 2). To be sure of the variability of the amount of RNA loaded for control and regeneratinglanes, we rehybridized 14 d after each filter with classIV P-tubulin and usedit as internal control, 7 d after nerve crush nerve crush becauseclass IV /3-tubulin did not appearto showmuch changein mRNA (% of control) (% of control) the expressionbetween control and regeneratingDRG, as shown ARG IP ARG in Table 1. Also, to be sure of the accuracy, quantification was KIFlA 75 70 70 performed both by the densitometryof autoradiogramsand by the KIFlB ND ND 51 useof the imagingplates, which are believedto have a wider range KIF2 65 68 50 of linearity. The quantitative resultsobtained in this manner for 7 KIF3A 6 kb 51 SO 54 d after nerve crush were very similar to those obtained for 14 d 4.4 kb 62 56 45 after nerve crush. Therefore, we did not detect the dramatic KIF3B 78 71 82 changesin the gene expressionof KIFlA, KIFlB, KIF2, KIF3A, KIF4 UD UD UD KIF3B, KIF4, or KIFS in regeneratinglumbar sensoryneurons. KIF5 55 64 58 Cytoplasmic dynein 142 117 ND Changes in mRNA expression of cytoplasmic dynein

The Northern blots such as those shown in Figures 2 and 3 were quantified by It hasbeen suggested that retrograde transport playsa role during densitometly of the autoradiograms (ARG) or by imaging plates (IP) for KIFlA, regeneration (Bisby, 1984; DiStefano and Curtis, 1994). There- KIFlB, KIIQ., KIF3A, KIF3B, KIF4, KIF5, and cytoplasmic dynein. The results were fore, we next examined the change in mRNA expressionof expressed as the percent of mRNA expression in regenerating DRG compared with control. For quantitation of mRNA expressed in DRG 7 d after nerve crush, the cytoplasmic dynein in regenerating mice. As shown in Figure 3 values were normalized with that of the class IV P-tubulin. For quantitation of and Table 2, there was a slight increase (up to 140%) in its mRNA expressed in DRG 14 d after nerve crush, the same amount of RNAs determined by optical density measurement were compared. ND, not determined; expression.Therefore, the mRNA expressionof cytoplasmicdy- UD, under detection limit in both control and regenerating DRG. nein wasregulated differently from KIFs, and the increasemay be 34 J. Neurosci., January 1, 1996, 76(1):31-35 Takemura et al. . KIF Expression and Regeneration

would lead to a high level of activity during both embryonic and CD regenerativeoutgrowth of axons.However, in the rat sensoryaxon the total amount of fast transport was found to be either de- creasedor unchangedduring regeneration, and it has been dis- cussedthat the required increasein the membranetransport may be compensatedfor by the decreaseof the total surfacearea in the truncated axon (Bisby, 1978; Grafstein and McQuarrie, 1978; Grafstein and Forman, 1980). Therefore, during the embryonic growth of axons,production of a large amount of KIF2 and KIF4 regulatedat the transcriptional level may be necessaryand, during regeneration, control of growth may be more important; robust increaseof KIF2 and KIF4 regulated at the transcriptional level 4.4- * + may be unnecessaryor even undesirable.Altered requirementsof b-3 membranetransport may be dealt with by more subtle control, such as post-transcriptionalor post-translationalregulation. Our resultsalso indicated that in addition to KIF2 and KIF4, the mRNA expressionof other KIFs, i.e., KIFlA, KIFlB, KIF3A, KIF3B, and KIFS, did not showmarked changesduring regener- TUB ation. Comparisonbased on the amount of total RNA or classIV P-tubulin expressionindicated that the gene expressionof KIFs Fi,qure 3. Northern hybridization autoradiograms of mouse DRG RNA examinedwas unchangedor slightly downregulatedduring regen- probed with cytoplasmic dynein. Seven days after nerve crush, RNA was eration. We do not have the information on whether the total isolated from L4 and L5 DRG. The same amount of total RNA (10 urr) RNA synthesisin DRG changesduring regeneration. Even if we was loaded for control (c) and regenerating (r) lanes. After hybridizati& assumethis, becauseof the inherent limitation in the accuracyof with cytoplasmic dynein probe, the same filter was rehybridized with class RNA quantification and becausethe degreeof downregulationwe IV p-tub&n probe. The numbers on the left indicate the size of the marker RNA in kilobases. observed was rather small compared with this limitation, it is difficult to distinguishthe two possibilities. During regeneration, certain proteins such as the enzymes related to the importance of retrograde transport during involved in synaptic transmissionare reduced (Grafstein and regeneration. McQuarrie, 1978; Grafstein and Forman, 1980; Koo et al., 1988), and certain proteins such as GAP43 are increased(Skene and DISCUSSION Willard, 1981a;Skene and Willard, 1981b;Hoffman, 1989;Skene, During axonal regeneration, the mRNA expressionpatterns of 1989;Tetzlaff et al., 1991).Because the gene expressionof none of neurofilament and tubulin change to that during development the KIFs examinedapparently was associatedwith suchchanges, (Hoffman et al., 1987;Hoffman and Cleveland, 1988;Miller et al., the altered transport was not likely to be accomplishedby the 1989;Wong and Oblinger, 1990;McKerracher et al., 1993).Thus, regulationof thesemotors at the transcriptional level. The activity the developmental program for cytoskeletal gene expressionis of motorsmay be regulatedby post-transcriptionalor post-trans- recapitulatedduring axonal regeneration.Our resultsshowed that lational regulations (Nixon, 1992; Sato-Yoshitake et al., 1992; the gene expressionof KIFs during regenerationdid not revert to Okada et al., 1995b),or the activity of motorsper se may not be that during development.The mRNA of KIF2 and KIF4, which related to the observed changesin the amount of transported are expressedabundantly during development,did not increase materials.Another possibility, althoughless likely, is that motors significantly. Therefore, in contrast to the gene expressionof that were not examined in this report are involved. KIF.5 appears major cytoskeletalelements, the developmentalprogram for an- to be an isotypeof kinesin heavy chain but is expressedexclusively terograde fast motors was not recapitulated during axonal in the brain (Aizawa et al., 1992;Niclas et al., 1994).Therefore, it regeneration. may be a gene product distinct from conventional kinesin heavy That the expressionof KIFs did not recapitulate the develop- chain. In addition, it is likely that one additional brain-specific mental program wassimilar to the resultsobtained by studieson isotype of kinesin heavy chain is present (Kato, 1990). the gene or expressionof MAPS during regeneration After nerve injury, there is a drastic responsein the cell body. (Woodhams et al., 1989; Oblinger et al., 1991; Svenssonand The rough endoplasmicreticulum is disorganized(“chromatoly- Aldskogius,1992; Fawcett et al., 1994).In thosestudies, it was the sis”) asfirst describedby Nisslin 1892(Grafstein and McQuarrie, general finding that the expressionof mRNA and protein of 1978).Free polyribosomesincrease concomitantly. The pattern of MAPS during regenerationretains the adult pattern. Therefore, it the gene expressionis altered. Retrograde transport is generally islikely that there is somedifference in the axon outgrowth during increased,and it hasbeen suggestedthat the retrograde transport developmentand regeneration.In both embryonic and regenera- is involved in conveying the signalfor thesechanges (Bisby, 1984). tive outgrowth of axon, induction of specific isotypes of tubulin In addition, recent evidence suggeststhat the retrograde trans- may be a prerequisite(Hoffman and Cleveland,1988; Miller et al., ports of neurotrophic factors are increasedduring regeneration 1989;McKerracher et al., 1993),but that of MAPS and KIFs may and play critical roles in peripheral nerve regeneration (Curtis et not be. In the caseof KIFs, the amount of KIF2 expressedin the al., 1993, 1994;DiStefano and Curtis, 1994). Our result that the adult may be satisfactoryalready for the requirement. It hasbeen retrograde transport motor isslightly upregulatedmay be relevant pointed out that the axon outgrowth during regenerationis slower to this aspect. Equally probable is that the increasedamount of and lessvigorous than the embryonic outgrowth (Fawcett et al., cytoplasmicdynein is necessaryto turn over the increasedamount 1994). One would expect that the addition of new membrane of cytoplasmicdebris. Takemura et al. l KIF Expression and Regeneration J. Neurosci., January 1, 1996, 16(1):31-35 35

The initial purpose of this study was to distinguish the roles of McKerracher L, Essagian C, Aguayo AJ (1993) Marked increase in /3-tu- different members of KIFs during regeneration, Contrary to our bulin mRNA expression during regeneration of axotomized retinal ganglion cells in adult mammals. J Neurosci 13:5294-5300. expectation, the expression of mRNA for any of the KIFs exam- Miller FD, Tetzlaff W, Bisby MA, Fawcett JW, Milner RJ (1989) Rapid ined, i.e., KIFlA, KIFlB, KIF2, KIF3A, KIF3B, KIF4, or KIF5, induction of the major embryonic cu-tubulin mRNA, Tal, during nerve did not change dramatically compared with others. Further stud- regeneration in adult rats. J Neurosci 9:1452-1463. ies will be necessary to determine separately the role of each Nangaku M, Sato-Yoshitake R, Okada Y, Noda Y, Takemura R, member of the KIFs in vivo. Yamazaki H, Hirokawa N (1994) KIFlB, a novel microtubule plus-end directed monomeric motor protein for transport of mitochondria. Cell REFERENCES 79:1209-1220. Aizawa H, Sekine Y, Takemura R, Zhang Z, Nangaku M, Hirokawa N Niclas J, Navone F, Horn-Booher N, Vale RD (1994) Cloning and local- (1992) Kinesin family in murine central nervous system. J Cell Biol ization of a conventional kinesin motor expressed exclusively in neu- 119:1287-1296. rons. Neuron 12:1059-1072. Bisby MA (1978) Fast axonal transport of labeled protein in sensory Nixon RA (1992) Slow axonal transport. Curr Opin Cell Biol 4:8-14. axons during regeneration. Exp Nemo1 61:281-300. Noda Y, Sato-Yoshitake R, Kondo S, Nangaku M, Hirokawa N (1995) Bisby MA (1984) Retrograde axonal transport and nerve regeneration. KIF2 is a new microtubule-based anterograde motor that transports In: Axonal transport in neuronal growth and regeneration (Elam JS, membranous organelles distinct from those carried by KHC or Cancalon P, eds), pp 45-67. New York: Plenum. KIF3A/B. J Cell Biol 129:157-167. Brady ST (1985) A novel brain ATPase with properties expected for the Oblinger MM, Argasinski A, Wong J, Kosik KS (1991) Tau gene expres- fast axonal transport motor. Nature 317:73-75. sion in rat sensory neurons during development and regeneration. J Chomczynski P, Sacchi N (1987) Single-step RNA isolation by acid gua- Neurosci 11:2453-2459. nidinium thiocyanate-phenol-chloroform extraction, Anal Biochem Okada Y, Yamazaki H, Sekine Y, Hirokawa N (1995a) The neuron 162:156-159. specific kinesin superfamily protein KIFlA is a unique monomeric Curtis R, Adryan KM, Zhu Y, Harkness PJ, Lindsay RM, DiStefano PS motor for the anterograde axonal transport of synaptic vesicle precur- (1993) Retrograde axonal transport of ciliary neurotrophic factor is sors. Cell 81:769-780. increased by peripheral nerve in&ny. Nature 365:253-255. Okada Y, Sato-Yoshitake R, Hirokawa N (1995b) The activation of Curtis R, Scherer SS. Somocrvi R. Adrvan KM. IO NY, Zhu Y. Lindsav protein kinase A pathway selectively inhibits anterograde axonal trans- RM, DiStefano PS (1994rRetrograde axonal iransport of LIF is in- port of vesicles but not mitochondria transport or retrograde transport creased by peripheral nerve injury: correlation with increased LIF in vivo. J Neurosci 15:3053-3064. expression in distal nerve. Neuron 12:191-204. Paschal BM, Shpetner HS, Valee RB (1987) MAPlC is a microtubule- DiStefano PS, Curtis R (1994) Receptor mediated retrograde axonal activated ATPase which translocates in vitro and has transport of neurotrophic factors is increased after peripheral_ _ nerve injury. Prog Brain Res 103:35-42. dynein-like properties. J Cell Biol 105:1273-1282. Fawcett JW. Mathews G. Housden E. Goedert M. Matus A 11994) Sato-Yoshitake R, Yorifuji H, Inagaki M, Hirokawa N (1992) The phos- Regenerating sciatic ne’rve axons contain the adult rather than the phorylation of kinesin regulates its binding to synaptic vesicles. J Biol embryonic pattern of microtubule associated proteins. Neuroscience Chem 26123930-23936. 61:789-804. Sekine Y, Okada Y, Noda Y, Kondo S, Aizawa H, Takemura R, Hirokawa Grafstein B, Forman DS (1980) Intracellular transport in neurons. N (1994) A novel microtubule-based motor protein (KIF4) for or- Physiol Rev 1167-1283. ganelle transports, whose expression is regulated developmentally. J Grafstein B, McQuarrie IG (1978) Role of the nerve cell body in axonal Cell Biol 127:187-201. regeneration. In: Neuronal nlasticitv (Cotman CW, ed),I DD. . 155-195. Skene JHP (1989) Axonal growth-associated proteins. Annu Rev Neuro- New York: Raven. _ . \ sci 12:127-156. Hirokawa N (1993) Axonal transport and . Curr Opin Neu- Skene JHP, Willard M (1981a) Changes in axonally transported proteins robiol 31724-731. during axon regeneration in toad retinal ganglion cells. J Cell Biol Hoffman PN (1989) Expression of GAP-43, a rapidly transported 89:86-95. growth-associated protein, and class II beta tubulin, a slowly trans- Skene JHP, Willard M (1981b) Axonally transported proteins associated ported cytoskeletal protein, are coordinated in regenerating neurons. J with axon growth in rabbit central and peripheral nervous systems. J Neurosci 9:893-897. Cell Biol 89:96-103. Hoffman PN, Cleveland DW (1988) Neurofilament and tubulin expres- Svensson M, Aldskogius H (1992) The effect of axon injury on microtu- sion recapitulates the developmental program during axonal regenera- bule-associated proteins MAP2, 3, and 5 in the hypoglossal nucleus of tion: induction of a specific P-tubulin isotype. Proc Nat1 Acad Sci USA the adult rat. J Neurocytol 21:222-231. 85:4530-4533. Tetzlaff W, Alexander SW, Miller FD, Bisby MA (1991) Response of Hoffman PN, Cleveland DW, Griffin JW, Landes PW, Cowan NJ, Price facial and rubrospinal neurons to axotomy: changes in mRNA expres- DL (1987) Neurofilament : a major determinant of sion for cytoskeletal proteins and GAP-43. J Neurosci 11:2528-2544. axonal caliber. Proc Nat1 Acad Sci USA 84:3472-3476. Vale RD, Reese TS, Sheetz MS (1985) Identification of a novel force- Kato K (1990) A collection of cDNA clones with specific expression generating protein, kinesin, involved in microtubule-based motility. Cell patterns in mouse brain. Eur J Neurosci 2:704-711. Kondo S, Sato-Yoshitake R, Noda Y, Aizawa H, Nakata T, Matsuura Y, 42:39-50. Hirokawa N (1994) KIF3 is a new microtubules based fast anterograde Wong J, Oblinger MM (1990) A comparison of peripheral and central motor in the nerve axons. J Cell Biol 125:1095-1107. axotomy effects on neurofilament and tubulin gene expression in rat Koo EH, Hoffman PN, Price DL (1988) Levels of neurotransmitter and dorsal root ganglion neurons. J Neurosci 10:2215-2222. cytoskeletal protein mRNAs during nerve regeneration in sympathetic Woodhams PL, Calvert R, Dunnett SB (1989) Monoclonal antibody GlO ganglia. Brain Res 449:361-363. against microtubule-associated protein lx distinguishes between grow- Laskey R (1980) The use of intensifying screens or organic scintillators ing and regenerating axons. Neuroscience 28:49-59. for visualizing radioactive molecules resolved by gel electrophoresis. Yamazaki H, Nakata T, Okada Y, Hirokawa N (1995) KIF3A/B, a new Methods Enzymol 65:363-371. heterodimer that works as microtubule plus-directed motor for fast Lewis SA, Cowan NJ (1985) Genetics, evolution, and expression of the axonal transport. J Cell Biol 130:1387-1399. 68,000-mol-wt neurofilament orotein: isolation of a cloned cDNA Zhang Z, Tanaka Y, Nonaka S, Aizawa H, Kawasaki H, Nakata T, probe. J Cell Biol 100:843-856 Hirokawa N (1993) The primary structure of rat brain (cytoplasmic) Lewis SA, Lee MG-S, Cowan NJ (1985) Five mouse tubulin isotypes and dynein heavy chain, a cytoplasmic motor enzyme. Proc Nat1 Acad Sci their regulated expression during development. J Cell Biol 101:852-861. USA 90:7928-7932.